A data-driven approach links microglia to pathology and prognosis in amyotrophic lateral sclerosis

Abstract

Amyotrophic lateral sclerosis (ALS) is a devastating neurodegenerative disease that lacks a predictive and broadly applicable biomarker. Continued focus on mutation-specific upstream mechanisms has yet to predict disease progression in the clinic. Utilising cellular pathology common to the majority of ALS patients, we implemented an objective transcriptome-driven approach to develop noninvasive prognostic biomarkers for disease progression. Genes expressed in laser captured motor neurons in direct correlation (Spearman rank correlation, p < 0.01) with counts of neuropathology were developed into co-expression network modules. Screening modules using three gene sets representing rate of disease progression and upstream genetic association with ALS led to the prioritisation of a single module enriched for immune response to motor neuron degeneration. Genes in the network module are important for microglial activation and predict disease progression in genetically heterogeneous ALS cohorts: Expression of three genes in peripheral lymphocytes - LILRA2, ITGB2 and CEBPD - differentiate patients with rapid and slowly progressive disease, suggesting promise as a blood-derived biomarker. TREM2 is a member of the network module and the level of soluble TREM2 protein in cerebrospinal fluid is shown to predict survival when measured in late stage disease (Spearman rank correlation, p = 0.01). Our data-driven systems approach has, for the first time, directly linked microglia to the development of motor neuron pathology. LILRA2, ITGB2 and CEBPD represent peripherally accessible candidate biomarkers and TREM2 provides a broadly applicable therapeutic target for ALS.

Keywords: Amyotrophic lateral sclerosis; Biomarkers; Microglia; Neuropathology; TREM2; Transciptome.

Fig. 1

Data-driven discovery workflow. Using anterior horn tissue, RNA transcript expression was measured from isolated motor neurons, and counts of p62-positive cytoplasmic inclusions within motor neurons were conducted. RNA expression and pathology counts from the same patients were correlated by Spearman’s rank correlation to identify 83 transcripts (a). Pathology correlated transcripts seeded co-expressed networks. The resulting combined network was developed into tightly co-expressing modules using weighted gene co-expression analysis (WGCNA) (b). Modules were prioritised using enrichment with independently curated gene lists related to ALS rate of progression and ALS genetic susceptibility. The two top scoring modules were enriched for neuronal and immune function respectively. MN = motor neuron, LB = lymphoblastoid (c). The immune module was selected for use as a biomarker in peripheral tissue and additional non-tissue specific genes were added. Components of the immune module were assessed by mRNA and protein quantification for predictive value in blood and cerebrospinal fluid (CSF) (d)

Fig. 2

Prioritisation and functional enrichment of genes modules. Gene co-expression modules associated with ALS neuropathology were identified using WGCNA; modules are numbered 1–82. Modules were tested for enrichment with three assessment gene sets curated to represent rate of progression in motor neurons (a) and lymphoblastoid cells (b), and upstream genetic association (c) with ALS. -log (p-value) refers to the p-value for enrichment of the corresponding module number with the relevant assessment set as calculated by Fisher’s exact test. Only modules significantly enriched with each assessment set (p-value <0.05) are plotted with respective p-values

Fig. 3

Construction of the immune network independent of cell type by addition of globally co-expressed genes and protein-protein interacting partners. The immune network module (module 27) contained 65 genes which was expanded to 77 genes by addition of globally co-expressed genes and protein-protein interacting partners. Each gene is represented by a node and is labelled with its HUGO identifier. Genes originating from module 27 are arranged on the left-side of the diagram; genes identified as globally co-expressed or protein-protein interacting partners are arranged on the right-side of the diagram. Relationships between genes are represented as edges between nodes, either global co-expression (purple) or protein-protein interaction (pink). Only genes with edges reaching statistical significance are shown. CEBPD, LILRA2 and ITGB2 (blue nodes), represent a proposed blood-based biomarker; TREM2 (red node) protein measured in CSF correlates with disease duration in selected patients

Fig. 4

Measurement of soluble TREM2 in CSF from ALS patients and controls. Soluble TREM2 levels were measured by ELISA in CSF from ALS patients (n = 46) and controls (n = 20) who were age and sex matched. Levels of soluble TREM2 are significantly higher in ALS patients compared to controls (Mann–Whitney, p < 0.05) (a). Stage of ALS at the time of sample was determined by the time from onset to sample compared to time from onset to death (censored). Levels of soluble TREM2 are highest in early ALS (CSF sampled in <25th centile of disease course), intermediately raised in late (>75th centile of disease course) and lowest in controls (b). Error bars show standard error. Levels of soluble TREM2 are positively correlated with disease duration in late stage ALS (c). We suggest a model whereby CSF soluble TREM2 is elevated in early disease in all ALS patients but then gradually reduces. In certain patients levels remain relatively high reflecting a prolonged neuroprotective microglial activation which leads to slower disease progression (d)